Machine learning (ML) approaches have become commonplace in many technical fields, however, their exploration, adoption, and exploitation in the areas of computational fluid dynamics (CFD), fluid modeling, and aircraft design is still an area of emerging research. Promising initial efforts have demonstrated the methods' effectiveness in expanding the accuracy of Reynolds Averaged Navier Stokes (RANS) methodologies to flows for which such methods typically exhibit poor performance. While such methods have shown initial promise, a challenge remains in determining proper methods for fusing computational and experimental training data while satisfying physical constraints. Others have begun to apply similar techniques to multifidelity approximation of flows, CFD solver convergence acceleration, and reduced order modeling. Application of such techniques could hold promise in reducing computational expense of standard CFD approaches, reducing man-in-the-loop demands in mesh generation, providing for rapid aerodynamic estimates by providing a framework for assimilating large amounts of computational parametric data, expanding validity of physical models, and assisting in solver characterization, among other possibilities. This topic envisions supporting exploration of such methods and their application in CFD and aerodynamics methods for air vehicle design analysis. Some areas of particular interest are: (1) Development of ML-based or corrected turbulence models, (2) Convergence acceleration via ML approximations, (3) Development of ML based rapid aerodynamic prediction capability based on parametric computational simulations, (4) ML assisted grid generation, and (5) ML based characterization of solver performance.